Hallo,

I am using the function

gsl_sf_ellint_Pcomp( k, n, mode )


from the GSL and do not understand, why n > 1 is not allowed. Obviously, I get the following error for larger n.

gsl: ellint.c:546: ERROR: domain error
Default GSL error handler invoked.
Aborted

Nevertheless, if I use a simple trapezoidal quadrature rule for the evaluation of the integral, I get nice results. From the definition
of

Pi( k, n ) =
        int_0^pi/2 (1 + n * sin^2 theta)^{-1}
        sqrt(1 - k^2 sin^2 theta)^{-1} d theta,

I don't see any restriction on 'n'.

With best regards,

        Thomas


--

Thomas Rüberg
Institute of Structural Analysis
++43-(0)316-873-6183
www.ifb.tugraz.at/thomas
TU Graz, Austria


_______________________________________________
Bug-gsl mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/bug-gsl

Reply via email to